Effective forgery detection using DCT+SVD-based watermarking for region of interest in key frames of vision-based surveillance

Abstract

This paper proposes a hybrid watermarking scheme based on the discrete cosine transform DCT and singular value decomposition SVD to apply to the forgery detection for region of interest ROI in key frames of video surveillance. In the proposed DCT+SVD-based watermarking, the cover image is transformed from the RGB colour space into the YCbCr colour space to obtain the grey-level image. Next, DCT technique is applied to the grey-level image to obtain the frequency components. Finally, SVD is used on the obtained frequency components. Using the proposed DCT+SVD-based watermarking on ROI in key frames of video surveillance, the forged frames can be effectively detected to prevent illegal forgery. The proposed forgery detection can prevent the influence of false positive detection using SVD-based watermarking algorithms. Experimental results show that the proposed method has good performance.